Microsoft introduces real-time natural language processing technology that translates other languages using speaker’s own voice [VIDEO]

// November 9th, 2012 // Futuristic Technology

Microsoft introduced a real-time voice language translation service that is leaps and bounds beyond what we would have thought possible with natural language interfaces.  In the demonstration video, you see the translation machine understanding English words and transcribing them to text as he speaks.  Then they demo the words being translated into Mandarin Chinese (a remarkable feat given the differences, such as word ordering, between the two languages).  But that’s not all.  Next the machine translates the text into spoken Mandarin – using his own tones.  The accuracy of the translator is obviously much better than other language processors on the market.

The head of Microsoft Research explained the breakthrough that allowed this translation technology:

“Just over two years ago, researchers at Microsoft Research and the University of Toronto made another breakthrough. By using a technique called Deep Neural Networks, which is patterned after human brain behavior, researchers were able to train more discriminative and better speech recognizers than previous methods. We have been able to reduce the word error rate for speech by over 30% compared to previous methods. This means that rather than having one word in 4 or 5 incorrect, now the error rate is one word in 7 or 8.”

Check out the video demonstration below.  It’s 9 minutes long but very interesting.

 

Microsoft speech recognition breakthrough technology




« « Previous Article: World War Z – Status Watch – Latest News     » » Next Article: Is NASA planning to deploy a robotic “miner” on the moon supported by a floating space station?


Leave a Reply

You must be logged in to post a comment.

%d bloggers like this: